Créer un CV ATS à partir d'un JD

• •Informations d'identification Google Drive API

{
  "id": "JgekNtar5mXcPtGd",
  "meta": {
    "instanceId": "fca4fa0e5ae247e39cd954acbd193c81c66eeb044912b56566983f18237c0bf6",
    "templateCredsSetupCompleted": true
  },
  "name": "ATS resume maker accoding to JD",
  "tags": [],
  "nodes": [
    {
      "id": "7f200ac3-741e-4c81-af61-b798c317801e",
      "name": "HTML en PDF",
      "type": "@custom-js/n8n-nodes-pdf-toolkit.html2Pdf",
      "position": [
        1664,
        112
      ],
      "parameters": {
        "htmlInput": "={{ $json.cleanedResponse }}"
      },
      "credentials": {
        "customJsApi": {
          "id": "4QZZxZEUk18tr3si",
          "name": "CustomJS account"
        }
      },
      "typeVersion": 1
    },
    {
      "id": "d6b2bfdf-b0e3-4477-a340-d97235f0ed92",
      "name": "Téléverser un fichier",
      "type": "n8n-nodes-base.googleDrive",
      "position": [
        2160,
        112
      ],
      "parameters": {
        "name": "NEW_RESUNME",
        "driveId": {
          "__rl": true,
          "mode": "list",
          "value": "My Drive"
        },
        "options": {},
        "folderId": {
          "__rl": true,
          "mode": "list",
          "value": "1vzNpRjBe1ylcLmJ2TKl4TN40BAeri-HD",
          "cachedResultUrl": "https://drive.google.com/drive/folders/1vzNpRjBe1ylcLmJ2TKl4TN40BAeri-HD",
          "cachedResultName": "Resume"
        }
      },
      "credentials": {
        "googleDriveOAuth2Api": {
          "id": "rEUMr90NJv8PD8bc",
          "name": "Google Drive account"
        }
      },
      "typeVersion": 3
    },
    {
      "id": "c78b7099-9c77-4ade-92b3-0677c5675dd1",
      "name": "Déclencheur de formulaire",
      "type": "n8n-nodes-base.formTrigger",
      "position": [
        -1344,
        112
      ],
      "webhookId": "285117be-9acd-496f-8320-866d6124a9fe",
      "parameters": {
        "path": "resume-builder",
        "options": {},
        "formTitle": "resume builder",
        "formFields": {
          "values": [
            {
              "fieldLabel": "Paste Resume text",
              "placeholder": "Resume plain text"
            },
            {
              "fieldType": "file",
              "fieldLabel": "Upload Resume PDF",
              "multipleFiles": false,
              "requiredField": true,
              "acceptFileTypes": ".pdf"
            },
            {
              "fieldType": "file",
              "fieldLabel": "Upload JD PDF",
              "multipleFiles": false,
              "requiredField": true,
              "acceptFileTypes": ".pdf"
            }
          ]
        },
        "responseMode": "lastNode"
      },
      "typeVersion": 1
    },
    {
      "id": "45e99a17-500f-4d4d-9c95-42a28c586427",
      "name": "Extraction CV1",
      "type": "n8n-nodes-base.extractFromFile",
      "position": [
        -400,
        112
      ],
      "parameters": {
        "options": {},
        "operation": "pdf",
        "binaryPropertyName": "={{ Object.keys($binary)[0] }}"
      },
      "typeVersion": 1,
      "alwaysOutputData": false
    },
    {
      "id": "825fd050-286c-4d9b-982a-b2028709e11b",
      "name": "Traitement d'un fichier binaire1",
      "type": "n8n-nodes-base.code",
      "position": [
        -848,
        112
      ],
      "parameters": {
        "jsCode": "let results = [];\n\nfor (let item of items) {\n    if (item.binary) {\n        // If there's binary data in the item, process each binary file\n        for (let key in item.binary) {\n            // Use the key as the file name\n            let binaryKey = key.replace(/\\s/g, 'Upload_Resume_PDF'); // Replace spaces with underscores for the key\n            results.push({\n                json: {\n                    fileName: binaryKey\n                },\n                binary: {\n                    [binaryKey]: item.binary[key] // Use the modified key for the binary data\n                }\n            });\n        }\n    }\n}\n\nreturn results;"
      },
      "typeVersion": 2
    },
    {
      "id": "86c96480-6969-4ea3-a839-8f235eb5c10c",
      "name": "Fusion CV + Offre1",
      "type": "n8n-nodes-base.code",
      "position": [
        48,
        112
      ],
      "parameters": {
        "jsCode": "return [\n  {\n    json: {\n      merged: $json.text + \" \" + $item(1).$json[\"text\"]\n    }\n  }\n];\n"
      },
      "typeVersion": 2
    },
    {
      "id": "84779a64-ecb1-4bf9-b9c3-bd968b3b47cf",
      "name": "Personnaliser CV1",
      "type": "n8n-nodes-base.perplexity",
      "position": [
        528,
        112
      ],
      "parameters": {
        "model": "sonar-reasoning",
        "options": {},
        "messages": {
          "message": [
            {
              "role": "system",
              "content": "=System (paste exactly as-is):\n\nYou are a professional resume writer. Build a single, ATS-friendly resume by combining and aligning the candidate’s resume with the job description contained in:\n{{ $json.merged }}\n\nRules for content:\n\nTailor the resume to the JD while staying truthful to the candidate’s information. Prioritize skills/keywords the JD demands. Rephrase; do not invent facts.\n\nOmit sections that have no data instead of adding placeholders.\n\nPrefer the candidate’s details when resume and JD conflict; only adjust wording to match JD terminology.\n\nRules for layout & style:\n\nOutput one valid HTML document only (no Markdown, no code fences, no JSON, no citations, no comments, no <think> or similar tags, no text before or after).\n\nUse inline CSS inside <style> in the <head>; no external assets (fonts, scripts, images, icons, tables).\n\nPlain white background, black text, font: Arial, sans-serif.\n\nPage-friendly margins (32–40px), single-column layout, consistent spacing.\n\nName at top center, bold, ~26–30px. Contact line centered beneath, smaller.\n\nSection titles (Summary, Skills, Experience, Projects, Education, Certifications, Links) left-aligned, bold, ~18–20px, with clear top margin.\n\nUse <ul><li> bullet points for skills and achievements. In Experience/Projects, show role/company/location/dates, then concise accomplishment bullets (use metrics where present).\n\nKeep it clean and ATS-safe: no colors, graphics, or fancy layouts.\n\nOutput requirement (critical):\nReturn only:\n\n<!DOCTYPE html>\n<html>…full document with inline <style>…</html>"
            },
            {
              "content": "=Final User Prompt (Strict Version):\n\nYou are a professional resume writer.\nYour task is to generate a single ATS-friendly resume in plain HTML and CSS only.\n\nInput data (resume + job description) is provided here:\n{{ $json.merged }}\n\nMandatory Guidelines (must follow exactly):\n\nOutput only valid HTML and CSS wrapped inside <html>...</html>. No explanations, comments, or extra text outside the HTML.\n\nThe resume must be clean, minimal, and professional:\n\nWhite background, black text, single font (Arial, sans-serif).\n\nCandidate’s full name: top center, bold, font size 26–30px.\n\nContact information: below name, centered, smaller font.\n\nSection titles (Skills, Certifications, Projects, Experience, Education, etc.): bold, left-aligned, font size 18–20px.\n\nUse bullet points for lists (skills, certifications, responsibilities, achievements).\n\nMaintain consistent spacing (margin-top for sections, padding inside lists).\n\nNo colors, tables, or unnecessary design elements.\n\nResume must combine the resume and job description into a single tailored version that best fits the target role.\n\nThe output must be ATS-optimized (clean structure, clear hierarchy, no hidden elements).\n\nDo not include placeholders like <think>, comments, or explanations. Only deliver the final HTML + CSS resume."
            }
          ]
        },
        "simplify": true,
        "requestOptions": {}
      },
      "credentials": {
        "perplexityApi": {
          "id": "jfeGy5zFESdZPlAv",
          "name": "Perplexity account"
        }
      },
      "typeVersion": 1
    },
    {
      "id": "bda8efee-e4a6-4c00-86e9-dd9c8fd0ac07",
      "name": "Format HTML1",
      "type": "n8n-nodes-base.code",
      "position": [
        1184,
        112
      ],
      "parameters": {
        "jsCode": "// Get the response from Perplexity\nconst response = $json.message || $json.content || $json.text;\n\n// Extract content between <html> and </html> tags\nconst htmlMatch = response.match(/<html[\\s\\S]*?<\\/html>/i);\n\nif (htmlMatch) {\n  // Found HTML content - now remove all \\n characters\n  const cleanHtml = htmlMatch[0].replace(/\\n/g, '');\n  \n  return [{\n    json: {\n      cleanedResponse: cleanHtml,\n      originalResponse: response\n    }\n  }];\n} else {\n  // No HTML found, return original\n  return [{\n    json: {\n      cleanedResponse: response,\n      originalResponse: response,\n      error: \"No HTML tags found in response\"\n    }\n  }];\n}"
      },
      "typeVersion": 2
    },
    {
      "id": "d9a5f53e-6683-4311-9d3b-790097efbb88",
      "name": "Note adhésive",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -1536,
        -480
      ],
      "parameters": {
        "color": 3,
        "width": 480,
        "height": 784,
        "content": "## Form Trigger Node Summary\n\nThe **Form Trigger** node in the *ATS Resume Maker according to JD* workflow initiates the process by providing a web form for users to submit their resume (text or PDF) and a job description (JD) PDF.\n\n- **Purpose**: Collects user inputs to start the resume tailoring process.\n- **Key Functions**:\n  - Creates a form at the path *resume-builder* with the title *resume builder*.\n  - Captures:\n    1. Optional plain text resume input.\n    2. Required resume PDF upload (accepts *.pdf* only).\n    3. Required JD PDF upload (accepts *.pdf* only).\n  - Uses *lastNode* response mode to wait for the workflow to complete.\n- **Output**: Sends text and binary (PDF) data to the next node (*Process one binary file1*).\n- **Role**: Acts as the workflow’s entry point, enabling user interaction.\n\n*Note*: This node is critical for gathering the raw data needed to generate an ATS-friendly resume. For more on n8n nodes, see [n8n Docs](https://.docs.n8n.io/workflows/components/)."
      },
      "typeVersion": 1
    },
    {
      "id": "c4c9e712-f353-460c-91d2-848ab373309f",
      "name": "Note adhésive1",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -880,
        880
      ],
      "parameters": {},
      "typeVersion": 1
    },
    {
      "id": "5de48c59-3453-4083-a31e-676294205864",
      "name": "Note adhésive2",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -544,
        -480
      ],
      "parameters": {
        "color": 4,
        "width": 400,
        "height": 784,
        "content": "## Extracting resume1 Node Summary\n\nThe **Extracting resume1** node in the *ATS Resume Maker according to JD* workflow extracts text from the binary PDF files (resume and JD) processed by the *Process one binary file1* node.\n\n- **Purpose**: Converts PDF files into readable text for further processing.\n- **Key Functions**:\n  - Uses the *pdf* operation to extract text from the binary data.\n  - Dynamically selects the binary file using the expression `={{ Object.keys($binary)[0] }}` to target the first available binary key.\n- **Output**: Text extracted from the PDFs, passed to the *Merge Resume + JD1* node.\n- **Role**: Enables the workflow to access the content of the resume and JD for tailoring.\n\n*Note*: This node is crucial for transforming PDF inputs into text that can be merged and customized. See [n8n Docs](https://docs.n8n.io/workflows/components/) for more on n8n nodes."
      },
      "typeVersion": 1
    },
    {
      "id": "46541729-54be-45f5-ba91-a1b294e854f3",
      "name": "Note adhésive3",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -1040,
        -480
      ],
      "parameters": {
        "color": 2,
        "width": 480,
        "height": 784,
        "content": "## Process one binary file1 Node Summary\n\nThe **Process one binary file1** node in the *ATS Resume Maker according to JD* workflow processes the binary data (PDF files) submitted through the *Form Trigger* node.\n\n- **Purpose**: Organizes and renames binary files (resume and JD PDFs) for further processing.\n- **Key Functions**:\n  - Uses JavaScript to iterate through input items and identify binary data (PDFs).\n  - Replaces spaces in binary file keys with *Upload_Resume_PDF* for consistent naming.\n  - Outputs:\n    1. JSON with the modified *fileName*.\n    2. Binary data mapped to the new key (e.g., *Upload_Resume_PDF*).\n- **Output**: Structured binary data passed to the *Extracting resume1* node for text extraction.\n- **Role**: Ensures clean, standardized file naming for downstream nodes.\n\n*Note*: This node prepares user-uploaded PDFs for text extraction, maintaining workflow efficiency. See [n8n Docs](https://docs.n8n.io/workflows/components/) for more on n8n nodes."
      },
      "typeVersion": 1
    },
    {
      "id": "00a054b3-a973-418c-840d-0796cc288b98",
      "name": "Note adhésive4",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -128,
        -480
      ],
      "parameters": {
        "color": 6,
        "width": 400,
        "height": 784,
        "content": "## Merge Resume + JD1 Node Summary\n\nThe **Merge Resume + JD1** node in the *ATS Resume Maker according to JD* workflow combines the extracted text from the candidate’s resume and the job description (JD) into a single text string.\n\n- **Purpose**: Merges resume and JD text to prepare for tailored resume generation.\n- **Key Functions**:\n  - Uses JavaScript to concatenate the resume text (`$json.text`) and JD text (`$item(1).$json[\"text\"]`) with a space separator.\n  - Outputs a single JSON object with a *merged* field containing the combined text.\n- **Output**: A unified text string passed to the *Customize resume1* node for further processing.\n- **Role**: Ensures both resume and JD content are combined for ATS-friendly customization.\n\n*Note*: This node sets the stage for tailoring the resume by providing a single text input. See [n8n Docs](https://docs.n8n.io/workflows/components/) for more on n8n nodes.\n\n*Timestamp*: 02:37 PM IST, Wednesday, August 20, 2025"
      },
      "typeVersion": 1
    },
    {
      "id": "cc1cd2d7-3315-4897-bc56-8d0d2c6606b4",
      "name": "Note adhésive5",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        288,
        -480
      ],
      "parameters": {
        "color": 7,
        "width": 592,
        "height": 784,
        "content": "## AI Perplexity Node Summary\n\nThe **Customize resume1** node in the *ATS Resume Maker according to JD* workflow uses the *Perplexity AI* model to generate an ATS-friendly resume by tailoring the merged resume and job description (JD) text into a structured HTML format.\n\n- **Purpose**: Leverages *Perplexity AI* to create a customized, ATS-optimized resume based on the merged input.\n- **Key Functions**:\n  - Employs the *sonar-reasoning* model to process the combined text (`$json.merged`) from the previous node.\n  - *Perplexity AI* follows strict instructions to:\n    - Produce valid HTML with inline CSS (Arial, sans-serif, black text, white background).\n    - Structure the resume with the candidate’s name (bold, 26–30px, centered), contact info, and sections like Skills, Experience, and Education using bullet points.\n    - Prioritize JD keywords while rephrasing truthfully, avoiding invented facts.\n    - Ensure ATS compatibility by using a clean, single-column layout without colors or graphics.\n- **Output**: HTML resume content passed to the *HTML format1* node for further processing.\n- **Role**: Relies on *Perplexity AI* to intelligently align the resume with the JD, ensuring relevance and scannability.\n\n*Note*: The *Perplexity AI* model is central to tailoring the resume, making it both professional and ATS-friendly. See [n8n Docs](https://docs.n8n.io/workflows/components/) for more on n8n nodes.\n\n"
      },
      "typeVersion": 1
    },
    {
      "id": "e5c8c302-9207-46bb-9151-8dd1edfa445a",
      "name": "Note adhésive6",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        896,
        -480
      ],
      "parameters": {
        "width": 592,
        "height": 784,
        "content": "## HTML format1 Node Summary\n\nThe **HTML format1** node in the *ATS Resume Maker according to JD* workflow processes the HTML resume generated by the *Customize resume1* node to ensure it is clean and properly formatted.\n\n- **Purpose**: Cleans and formats the HTML output from the Perplexity AI model for downstream use.\n- **Key Functions**:\n  - Uses JavaScript to extract the content between `<html>` and `</html>` tags from the AI’s response (`$json.message`, `$json.content`, or `$json.text`).\n  - Removes all newline characters (`\\n`) from the HTML to create a compact, single-line string.\n  - Outputs a JSON object with:\n    - *cleanedResponse*: The cleaned HTML content.\n    - *originalResponse*: The original AI response.\n    - *error*: A message if no HTML tags are found.\n- **Output**: Cleaned HTML passed to the *HTML3* and *HTML to PDF* nodes for further processing.\n- **Role**: Ensures the HTML resume is properly formatted and free of unwanted characters for conversion to PDF.\n\n*Note*: This node refines the AI-generated HTML for consistency and compatibility with subsequent nodes. See [n8n Docs](https://docs.n8n.io/workflows/components/) for more on n8n nodes.\n\n*Timestamp*: 02:42 PM IST, Wednesday, August 20, 2025"
      },
      "typeVersion": 1
    },
    {
      "id": "b8d0caab-6392-4123-a180-d33665cd4c54",
      "name": "Note adhésive7",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        1184,
        640
      ],
      "parameters": {
        "color": 3,
        "width": 448,
        "height": 480,
        "content": "## Generate HTML Preview Node Summary\n\nThe **HTML3** node in the *ATS Resume Maker according to JD* workflow processes the cleaned HTML resume output from the *HTML format1* node for further handling or validation.\n\n- **Purpose**: Handles the formatted HTML resume to prepare it for potential display or additional processing.\n- **Key Functions**:\n  - Receives the cleaned HTML content (`$json.cleanedResponse`) from the *HTML format1* node.\n  - Uses the `n8n-nodes-base.html` node (version 1.2) to process the HTML, likely for rendering, validation, or storage purposes.\n- **Output**: The processed HTML, which may be used for visualization or debugging (specific output depends on workflow context).\n- **Role**: Acts as a processing step for the HTML resume, ensuring it’s ready for any additional use cases, such as display or logging.\n\n*Note*: This node ensures the HTML resume is properly handled before final output or conversion. See [n8n Docs](https://docs.n8n.io/workflows/components/) for more on n8n nodes.\n"
      },
      "typeVersion": 1
    },
    {
      "id": "dbe2e4c3-cb5d-4f04-84ed-8071cee8904b",
      "name": "Aperçu HTML",
      "type": "n8n-nodes-base.html",
      "position": [
        1344,
        480
      ],
      "parameters": {
        "html": "{{ $json.cleanedResponse }}"
      },
      "typeVersion": 1.2
    },
    {
      "id": "438498a1-ea92-4b2e-833d-76b05cc25139",
      "name": "Note adhésive8",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        1504,
        -480
      ],
      "parameters": {
        "color": 3,
        "width": 416,
        "height": 784,
        "content": "## HTML to PDF Node Summary\n\nThe **HTML to PDF** node in the *ATS Resume Maker according to JD* workflow converts the cleaned HTML resume from the *HTML format1* node into a PDF file.\n\n- **Purpose**: Generates a PDF version of the tailored, ATS-friendly resume.\n- **Key Functions**:\n  - Uses the `@custom-js/n8n-nodes-pdf-toolkit.html2Pdf` node (version 1) to convert the HTML content (`$json.cleanedResponse`) into a PDF.\n  - Relies on credentials for a *CustomJS account* to access the PDF conversion service.\n- **Output**: A binary PDF file passed to the *Upload file* node for storage.\n- **Role**: Transforms the HTML resume into a downloadable PDF format, completing the resume creation process.\n\n*Note*: This node finalizes the resume as a PDF, making it ready for user download or storage. See [n8n Docs](https://docs.n8n.io/workflows/components/) for more on n8n nodes.\n\n"
      },
      "typeVersion": 1
    },
    {
      "id": "28c9b45a-9508-459a-94a1-a183c8a14280",
      "name": "Note adhésive9",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        1936,
        -480
      ],
      "parameters": {
        "color": 5,
        "width": 480,
        "height": 784,
        "content": "## Upload file Node Summary\n\nThe **Upload file** node in the *ATS Resume Maker according to JD* workflow uploads the generated PDF resume to Google Drive.\n\n- **Purpose**: Stores the ATS-friendly resume PDF in a specified Google Drive folder.\n- **Key Functions**:\n  - Uses the `n8n-nodes-base.googleDrive` node (version 3) to upload the PDF file.\n  - Configured to save the file as *NEW_RESUNME* in the *Resume* folder (`1vzNpRjBe1ylcLmJ2TKl4TN40BAeri-HD`) on *My Drive*.\n  - Authenticates using a *Google Drive account* credential.\n- **Output**: Confirmation of the file upload to Google Drive, completing the workflow.\n- **Role**: Ensures the final resume PDF is saved to the user’s Google Drive for easy access.\n\n*Note*: This node finalizes the workflow by storing the tailored resume PDF in a designated cloud location. See [n8n Docs](https://docs.n8n.io/workflows/components/) for more on n8n nodes.\n\n"
      },
      "typeVersion": 1
    }
  ],
  "active": true,
  "pinData": {},
  "settings": {
    "executionOrder": "v1"
  },
  "versionId": "51d3f7e1-1c0e-41d2-9a2e-de31374df1e9",
  "connections": {
    "7f200ac3-741e-4c81-af61-b798c317801e": {
      "main": [
        [
          {
            "node": "d6b2bfdf-b0e3-4477-a340-d97235f0ed92",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "c78b7099-9c77-4ade-92b3-0677c5675dd1": {
      "main": [
        [
          {
            "node": "825fd050-286c-4d9b-982a-b2028709e11b",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "bda8efee-e4a6-4c00-86e9-dd9c8fd0ac07": {
      "main": [
        [
          {
            "node": "dbe2e4c3-cb5d-4f04-84ed-8071cee8904b",
            "type": "main",
            "index": 0
          },
          {
            "node": "7f200ac3-741e-4c81-af61-b798c317801e",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "84779a64-ecb1-4bf9-b9c3-bd968b3b47cf": {
      "main": [
        [
          {
            "node": "bda8efee-e4a6-4c00-86e9-dd9c8fd0ac07",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "45e99a17-500f-4d4d-9c95-42a28c586427": {
      "main": [
        [
          {
            "node": "86c96480-6969-4ea3-a839-8f235eb5c10c",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "86c96480-6969-4ea3-a839-8f235eb5c10c": {
      "main": [
        [
          {
            "node": "84779a64-ecb1-4bf9-b9c3-bd968b3b47cf",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "825fd050-286c-4d9b-982a-b2028709e11b": {
      "main": [
        [
          {
            "node": "45e99a17-500f-4d4d-9c95-42a28c586427",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  }
}